SEARCH · Search NASA

Results for “Model Optimization”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

SourceResearch connectionFrom yearThrough year

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 433 records · Page 24

All sources NASA NTRS DOE OSTI

DOE OSTI2025

ATTNChecker: Highly-Optimized Fault Tolerant Attention for Large Language Model Training

Large Language Models (LLMs) have demonstrated remarkable performance in various natural language processing tasks. However, the training of these models is computationally intensive and susceptible to faults, particularly in the attention mechanism, which is a critical component of transformer-based LLMs. In this paper, we investigate the impact of faults on LLM training, focusing on INF, NaN, and near-INF values in the computation results with systematic fault injection experiments. We observe the propagation patterns of these errors, which can trigger non-trainable states in the model and disrupt training, forcing the procedure to load from checkpoints. To mitigate the impact of these faults, we propose ATTNChecker, the first Algorithm-Based Fault Tolerance (ABFT) technique tailored for the attention mechanism in LLMs. ATTNChecker is designed based on fault propagation patterns of LLM and incorporates performance optimization to adapt to both system reliability and model vulnerability while providing lightweight protection for fast LLM training. Evaluations on four LLMs show that ATTNChecker on average incurs on average 7% overhead on training while detecting and correcting all extreme errors. Compared with the state-of-the-art checkpoint/restore approach, ATTNChecker reduces recovery overhead by up to 49×.

Liang, Yuhang [University of Alabama - Birmingham]↗

DOE OSTI2024

Bayesian Optimization for Anything (BOA): An open-source framework for accessible, user-friendly Bayesian optimization

We introduce Bayesian Optimization for Anything (BOA), a high-level Bayesian Optimization (BO) framework and model wrapping toolkit, which presents a novel approach to simplifying BO, with the goal of making it more accessible and user-friendly, particularly for those with limited expertise in the field. BOA addresses common barriers in implementing BO, focusing on ease of use, reducing the need for deep domain knowledge, and cutting down on extensive coding requirements. A notable feature of BOA is its language-agnostic architecture, which facilitates broader application in various fields and to a wider audience. We showcase BOA's application through three examples: a high-dimensional optimization with parameters of the SWAT+ watershed model, a highly parallelized optimization of this intrinsically non-parallel model, and a multi-objective optimization of the FETCH Tree-Crown Hydrodynamics model. Furthermore, these test cases illustrate BOA's effectiveness in addressing complex optimization challenges in diverse scenarios.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2025

Compressing Vision Transformers in Geospatial Transfer Learning with Manifold-Constrained Optimization

Deploying geospatial foundation models on resource-constrained edge devices demands compact architectures that maintain high downstream performance. However, their large parameter counts and the accuracy loss often induced by compression limit practical adoption.In this work, we leverage manifold-constrained optimization framework DLRT to compress large vision transformer–based geospatial foundation models during transfer learning. By enforcing structured low-dimensional parameterizations aligned with downstream objectives, this approach achieves strong compression while preserving task-specific accuracy. We show that the method outperforms of-the-shelf low-rank methods as LoRA. Experiments on diverse geospatial benchmarks confirm substantial parameter reduction with minimal accuracy loss, enabling high-performing, on-device geospatial models.

Snyder, Thomas [Yale University]↗

DOE OSTI2025

Sparse measurement medical CT reconstruction using multi-fused block matching denoising priors

A major challenge for medical X-ray CT imaging is reducing the number of X-ray projections to lower radiation dosage and reduce scan times without compromising image quality. However these under-determined inverse imaging problems rely on the formulation of an expressive prior model to constrain the solution space while remaining computationally tractable. Traditional analytical reconstruction methods like Filtered Back Projection (FBP) often fail with sparse measurements, producing artifacts due to their reliance on the Shannon-Nyquist Sampling Theorem. Consensus Equilibrium, which is a generalization of Plug and Play, is a recent advancement in Model-Based Iterative Reconstruction (MBIR), has facilitated the use of multiple denoisers are prior models in an optimization free framework to capture complex, non-linear prior information. However, 3D prior modelling in a Plug and Play approach for volumetric image reconstruction requires long processing time due to high computing requirement. Instead of directly using a 3D prior, this work proposes a BM3D Multi Slice Fusion (BM3D-MSF) prior that uses multiple 2D image denoisers fused to act as a fully 3D prior model in Plug and Play reconstruction approach. Our approach does not require training and are thus able to circumvent ethical issues related with patient training data and are readily deployable in varying noise and measurement sparsity levels. In addition, reconstruction with the BM3D-MSF prior achieves similar reconstruction image quality as fully 3D image priors, but with significantly reduced computational complexity. We test our method on clinical CT data and demonstrate that our approach improves reconstructed image quality.

Hossain, Maliha [ORNL]↗

DOE OSTI2025

Machine learning-driven design and self-sensing capabilities of automotive bumper lattices for adaptive impact response

We present a novel approach to design an automotive bumper energy absorber using carbon fiber reinforced polymer composites, optimized to meet conflicting performance requirements for two distinct impact scenarios. The design must satisfy both a low-speed (2.5 mph) pendulum intrusion test, simulating vehicle-to-vehicle collisions, and a high-speed (25 mph) leg flexion test, replicating pedestrian impacts. These tests demand opposing deformation characteristics: high flexibility (deformation < 85 mm) for the former and high stiffness (deformation < 22 mm) for the latter. To address these contradictory requirements, we developed a machine learning (ML) framework for inverse optimization of lattice designs and material selection. Unlike traditional iterative design processes, our ML model directly outputs optimal design parameters and material choices based on target performance inputs. The energy absorber was fabricated using advanced additive manufacturing techniques, including extrusion deposition and digital light processing. The integration of carbon fibers provides multifunctionality to the bumper structure, enabling self-sensing capabilities through changes in electrical resistivity under compression. This electrical response demonstrates high repeatability under multiple cycles at 2% compression and exhibits distinct signatures during crack formation under high deformation. This research offers adaptive performance through innovative design methodologies and smart material integration. The approach has potential applications in various fields requiring adaptive energy absorption and real-time structural health monitoring.

Chawla, Komal [ORNL] (ORCID:0000000190327565)↗

DOE OSTI2024

Datasets for Widespread Residential Space Heating Electrification in Texas

In this experiment, we explore long term patterns in electricity demand driven by the dual effects of full electrification of space heating in Texas (by adoption of electric heat pumps), and climate change. We use a predictive model of electricity demand, climate projections, and an open source nodal power system (DC Optimal Power Flow) model of the Electric Reliability Council of Texas (ERCOT) system. Heat pumps are a more energy efficient way of providing space heating and cooling in homes. We attempt to exhaustively investigate the impacts of full residential space heating electrification by adoption of heat pumps for the segment of Texas households that currently rely on fossil fuels (about 40%), while simultaneously incorporating climate change meteorological variables. We explore a range of scenarios of heat pump efficiency and climate uncertainty over a long period of future years (2020-2099). In total, the simulation experiment generates 1,280 simulation years of hourly data. We report and analyze results in form of impacts on residential load, total load, peak load, seasonality of peaking, and reliability measured by occurrence and frequency of loss of load events. While the experiment is for ERCOT, the insights and approach can be applied to other regions. The results from the analysis can inform system planners on a range of potential capacity requirements/ reliability implications and/or risks of full space heating electrification via the adoption of electric heat pumps, given the uncertainty in the scenarios/ climate futures. The dataset includes model output for residential, non residential and total load, and the results from the GO ERCOT model runs for 4 RCP Scenarios (RCP 4.5 Cooler, RCP 4.5 Hotter, RCP 8.5 Cooler, RCP 8.5 Hotter), 4 Heating electrification Scenarios (Base , Standard Efficiency HP, High Efficiency HP, Ultra-High Efficiency HP) over 80 years (2020-2099). The metrological variables at BA scale were weighted weighted using population projections consistent with the SSP3 scenario.

Climate Change↗

DOE OSTI2025

Autonomous alloy composition optimization using molecular dynamics guided by a large language model

Here, we present an autonomous materials discovery framework that couples a large language model (LLM) with molecular dynamics (MD) simulations to optimize Fe–Cr–Mn alloy compositions for tensile strength. Starting from six distinct compositions, the LLM operated as an intelligent agent, iteratively proposing changes based on prior simulation results and constraints. Over 50 iterations per case, the LLM adaptively explored the composition space, identifying high-strength regions, not easily accessible by conventional methods. The highest strength, 18.7 GPa, was achieved with Fe 71 Cr 25 Mn 4 composition, identified from a Fe 75 Cr 20 Mn 5 starting point. The LLM autonomously adjusted its strategy in real time, demonstrating closed-loop decision-making using commodity hardware. This approach showcases the potential of LLMs as scientific co-pilots, capable of accelerating materials discovery and generalizable to other domains like biology and drug design.

Autonomy↗

DOE OSTI2025

Predicting the Evolution of Shallow Cumulus Clouds With a Lotka‐Volterra Like Model

Abstract In numerical weather prediction and climate models, boundary‐layer clouds are controlled by a wide range of subgrid‐scale processes. However, understanding the nature of these processes and their role in the evolution of the cloud size distribution as a whole has been elusive. To address this issue, we adopt a novel empirical framework from the field of population dynamics to model the evolution of cloud size statistics by using the shallow cumulus properties obtained from a large‐eddy simulation (LES). Our approach involves representing the cloud size distribution and the total cloud area using a revised Lotka‐Volterra model and ridge linear model, respectively. The physical interpretation of the total cloud area and coefficients obtained from the optimization of the models reveals three stages probably interpreted by dominant processes: the formation of new clouds, the growth of single clouds, and a steady state with organized transitions involving the growth and decay of multiple clouds. Furthermore, we showcase the potential of this framework to serve as a component of scale‐aware parameterizations of shallow‐convective clouds in atmospheric models.

54 ENVIRONMENTAL SCIENCES↗

DOE OSTI2024

Impact of Higher Fidelity Design Iterations on Critical System Criteria

Nuclear criticality experiments are effective at informing the performance of nuclear data libraries across many applications. This work explores the implications of refining critical experiment MCNP models from their low fidelity optimization phase to penultimate neutronic models. Specifically, this work is focused on two series of plutonium fueled experiments funded through internal programs at Los Alamos National Laboratory building off previous efforts under the EUCLID (Experiments Underpinned by Computational Learning for Improvements in Nuclear Data) collaboration. Thales, the first of the two collaborations, is a fast spectrum Ta-reflected plutonium experiment to support operations at PF-4. The second experiment are twin configurations designed to target the intermediate energy cross sections in 239 Pu. Motivation for this experiment stems from the PARallel Approach of Differential and InteGral Measurements (PARADIGM) collaboration which hopes to achieve a significant reduction in 239 Pu cross section uncertainties in the intermediate region.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2024

Uncertainty quantification of mass models using ensemble Bayesian model averaging

Developments in the description of the masses of atomic nuclei have led to various nuclear mass models that provide predictions for masses across the whole chart of nuclides. These mass models play an important role in understanding the synthesis of heavy elements in the rapid neutron capture ( r ) process. However, it is still a challenging task to estimate the size of uncertainty associated with the predictions of each mass model. In this work, a method called ensemble Bayesian model averaging (EBMA) is introduced to quantify the uncertainty of one-neutron separation energies (S 1 n ) which are directly relevant in the calculations of r -process observables. Here, this Bayesian method provides a natural way to perform model averaging, selection, and uncertainty quantification, by combining the mass models as a mixture of normal distributions whose parameters are optimized against the experimental data, employing the Markov chain Monte Carlo method using the no-u-turn sampler. The EBMA model optimized with all the experimental S 1 n from the AME2003 nuclides are shown to provide reliable uncertainty estimates when tested with the new data in the AME2020.

73 NUCLEAR PHYSICS AND RADIATION PHYSICS↗

DOE OSTI2026

Good practices for documenting AI-based studies on energy and buildings

Artificial intelligence has transformed building science research over the past decade, with applications spanning energy modeling, energy prediction, HVAC optimization and controls, fault detection, and occupancy modeling. However, many studies lack adequate documentation of datasets, algorithms, training procedures, and validation methods. Building science research faces additional challenges including inconsistent evaluation metrics, limited generalizability across building types, climates, and significant gaps between experimental studies and deployed systems. This communication provides practical guidance for good practices in documenting and publishing AI-based research following established standards from the computer science and machine learning communities. By adopting frameworks such as Datasheets for Datasets, Model Cards, and standardized reproducibility checklists, researchers can ensure their work meets the rigorous documentation standards necessary for reproducible, comparable, and impactful building science research.

Hong, Tianzhen [Lawrence Berkeley National Laborat↗

DOE OSTI2025

Record acceleration of the two-dimensional Ising model using a high-performance wafer-scale engine

The versatility and wide-ranging applicability of the Ising model, originally introduced to study phase transitions in magnetic materials, have made it a cornerstone in statistical physics and a valuable tool for evaluating the performance of emerging computer hardware. Here, we present a novel implementation of the two-dimensional Ising model on Cerebras Wafer-Scale Engine (WSE) – a revolutionary processor that is opening new frontiers in computing. In our deployment of the checkerboard algorithm, we optimized the Ising model to take advantage of the unique WSE architecture. Specifically, we employed a compressed bit representation storing 16 spins on each int16 word, and efficiently distributed the spins over the processing units enabling seamless weak scaling and limiting communications to only immediate neighboring units. Our implementation can handle up to 754 simulations in parallel, achieving an aggregate of over 61.8 trillion flip attempts per second for Ising models with up to 200 million spins. This represents a gain of up to 148 times over previously reported single-devices with a highly optimized implementation on NVIDIA V100 and up to 88 times in productivity compared to NVIDIA H100. Our findings highlight the significant potential of the WSE in scientific computing, particularly in the field of materials modeling.

Ising model↗

DOE OSTI2025

Loss Landscape Analysis for Reliable Quantized ML Models for Scientific Sensing

In this paper, we propose a method to perform empirical analysis of the loss landscape of machine learning (ML) models. The method is applied to two ML models for scientific sensing, which necessitates quantization to be deployed and are subject to noise and perturbations due to experimental conditions. Our method allows assessing the robustness of ML models to such effects as a function of quantization precision and under different regularization techniques -- two crucial concerns that remained underexplored so far. By investigating the interplay between performance, efficiency, and robustness by means of loss landscape analysis, we both established a strong correlation between gently-shaped landscapes and robustness to input and weight perturbations and observed other intriguing and non-obvious phenomena. Our method allows a systematic exploration of such trade-offs a priori, i.e., without training and testing multiple models, leading to more efficient development workflows. This work also highlights the importance of incorporating robustness into the Pareto optimization of ML models, enabling more reliable and adaptive scientific sensing systems.

Baldi, Tommaso [Pisa, Scuola Normale Superiore]↗

DOE OSTI2025

Electromagnetic coil optimization for reduced Lorentz forces

Abstract The reduction of magnetic forces on electromagnetic coils is an important consideration in the design of high-field devices such as the stellarator or tokamak. Unfortunately, these forces may be too time-consuming to evaluate by conventional finite element modeling within an optimization loop. Although mutual forces can be computed rapidly by approximating large-bore coils as infinitely thin, this approximation does not hold for self-forces as it leads to an unphysical divergence. Recently, a novel reduced model for the self-field, self-force, and self-inductance of electromagnetic coils based on filamentary models was rigorously derived and demonstrated to be highly accurate and numerically efficient to evaluate (Hurwitz et al 2024 IEEE Trans. Magn. 60 7001614). In this paper, we present an implementation of the reduced self-force model employing automatic differentiation within the simsopt stellarator design software and use it in derivative-based coil optimization for a quasi-axisymmetric stellarator. We show that it is possible to significantly reduce point-wise forces throughout the coils, though this comes with trade-offs to fast particle losses and the minimum distance between coils and the plasma surface. The trade-off between magnetic forces and coil-surface distance is mediated by the minimum coil–coil distance for coils near the inboard side of the ‘bean’ cross-section of the plasma. The relationship between forces and fast particle losses is mediated by the normal field error. Coil forces can be lowered to a threshold with minimal deterioration to losses. Importantly, the magnet optimization approach here can be used also for tokamaks, other fusion concepts, and applications outside of fusion.

Hurwitz, Siena (ORCID:0000000166599659)↗

DOE OSTI2025

Fleet Algorithm Design for Pooled Rideshare: Integrating Human Factors, Simulation, and Optimization

This dissertation explores the study the integration of human factors modeling and rideshare fleet control algorithms. Pooled rideshare is a unique transportation mode offering that allows riders increased flexibility and accessibility over public transportation, and decreased cost relative to personal vehicles or traditional rideshare. Additionally, relative to personal vehicles, pooled rideshare offers reduced costs and options for those with difficulty obtaining transportation. Prior research in the space typically focused on modeling human behavior, or optimizing system performance, but a lack of integration of the concepts leads to unrealistic or underutilized outcomes. To tackle this problem, novel rideshare assignment, and repositioning strategies were designed and implemented in a simulation environment. Through a series of successive studies, improvements to current rideshare processes were identified, and beneficial outcomes for profitability, accessibility, and traffic were explored. Further, improved metrics to assess rideshare performance were designed and analyzed in the context of improved rideshare offerings. This research contributes to the field of transportation by tackling novel but pragmatic approaches to challenges facing the rideshare industry. Through the course of this dissertation, rideshares impacts on users, operators, and even regulators will be explored in detail. The justification behind the use of a simulation environment, a set of simulated regional models for testing, and the focus on realism and deployability is illustrated. The research identifies holes in potential markets for the use of both private, and public rideshare systems.

Paul, Joseph↗

DOE OSTI2024

Reduction of Methane Leaks through Corrosion Mitigation Pre-treatments for Pipelines with Field Applied Coatings

Corrosion of buried, coated steel pipelines transporting natural gas is a significant source of methane emissions, from pipeline venting required for maintenance and repairs and from pipeline leaks and incidents. Corrosion of steel under field applied coatings is an important safety concern for the pipeline industry. This project investigates the application of a field applied alloy over girth welds to mitigate external corrosion of buried coated steel pipelines. Various metallic coating options were considered, which were required to meet several criteria: (1) it must resist corrosion under open-circuit or mild cathodic protection conditions, (2) it must protect the substrate steel, and (3) it must not negatively affect the adhesion of the field coating. Finite element models and lab testing were performed of alloy coating compositions to identify promising alloy types underneath disbonded coatings. Polarization curves of coating alloys were generated to provide the boundary conditions for the COMSOL model to compute potential and current distributions around coated areas. Sacrificial and corrosion-resistant metal alloy coatings were evaluated and optimized using corrosion modeling and laboratory electrochemical testing, where aluminum alloy 5356 (5% Mg) and steel alloy B9 (9% Cr) were selected. Corrosion test coupons were designed and fabricated using thermal spray aluminum 5356 and welded B9 steel overlays on API 5L grade X42 line pipe steel. The corrosion test coupons, with simulated pipe coating damage, were tested in a laboratory soil box and a field pipeline site in Texas for 3-months. Corrosion test coupons were then tested for 6-months at field pipeline sites in Texas and Tennessee to quantify corrosion rates and performance of the aluminum and steel alloys under polyethylene tape and 2-part epoxy coatings, various coating holidays, and with and without cathodic protection.

03 NATURAL GAS↗

DOE OSTI2025

Systematic analysis of magnetic equilibrium reconstruction with eddy currents on LTX- β

Magnetic equilibrium reconstruction in the Lithium Tokamak Experiment-Beta (LTX-β) is complicated by strong eddy currents and toroidal asymmetries arising from its segmented, close-fitting conducting shell. In the earlier experiment LTX, these three-dimensional (3D) conductor effects significantly distorted diagnostic signals and challenged conventional axisymmetric reconstruction methods. In this work, we demonstrate that accurate plasma equilibria can be recovered on LTX-β by incorporating a small number of dominant eddy current modes derived from realistic conductor models. Using the open-source $\tt{TokaMaker}$ Grad–Shafranov solver, we reconstruct equilibria across a systematically selected set of LTX-β discharges and validate them against the legacy $\tt{PSI-Tri}$ hybrid 2D-3D code. Our results show that fully 2D $\tt{TokaMaker}$ reconstructions achieve significantly improved agreement with flux loop and Mirnov probe measurements, reducing total chi-squared fitting errors, especially during startup. Finally, we develop new hybrid 2D-3D $\tt{TokaMaker}$ reconstructions by integrating the $\tt{ThinCurr}$ 3D eddy current model, which we optimize for significant further reduction of chi-squared in most scenarios. These findings underscore the importance of realistic wall-current modeling in short-pulse tokamaks, and establish a physics-based reconstruction framework that is extensible to devices with complex passive structures and 3D wall interactions.

Grad–Shafranov solver↗

DOE OSTI2024

Development and Validation of Home Comfort System for Total Performance Deficiency/Fault Detection and Optimal Comfort Control

In this project, we developed and tested a learning-based home thermal model that facilitates the operation of a model predictive control (MPC)-based optimization agent and an automated fault detection and diagnosis (AFDD) agent. The home thermal model was constructed using a two-node resistor-capacitor model. Moreover, two accompanying parameter identification methods were introduced, least-squares and optimization. Based on the home thermal model, the MPC-based optimization agent was developed to optimize residential HVAC operation. Using two FDD methods, the AFDD agent was constructed to detect and diagnose two prevalent residential AC faults, airflow reduction and refrigerant undercharge. The home thermal model, along with the MPC-based optimization agent and AFDD agent, were tested at the Norman Test House, Miami Test House, Pacific Northwest National Laboratory (PNNL) Test House A, and PNNL Test House B. Finally, they were also field tested in nine demonstration homes with real occupants.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗